P1321PLASMATIC OSTEOPONTIN: A PREDICTOR OF VASCULAR ACCESS DYSFUNCTION IN HEMODIALYSIS PATIENTS

Abstract

Abstract Background and Aims Despite recommendations for a close follow-up of arteriovenous fistulas (AVF), thombosis rate reaches 10% each year. Conventional follow-up modalities remain a burden for hemodialysis patients. We tested the hypothesis that osteopontin (OPN), a pro-inflammatory molecule related to intimal hyperplasia, could be a biological marker of stenosis, and could thereby allow a screening of patients at risk for AVF dysfunction. Method Our prospective study (NCT03270358) compared the rate of plasmatic OPN between patients with a good-functioning AVF (control group, N= 39) and patients who required surgical or endovascular revision of their AVF because of a stenosis (stenosis group, N= 37). Blood samples were taken in the AVF, at the time of AVF cannulation in control patients, and at the time of AVF revision in stenosis patients. For each patient, blood from the contralateral arm was also analyzed as a between-subject control (paired t test). Cardiovascular risk factors, ongoing medications, and OPN rates (ELISA technique) were compared among the groups (uni- and multivariate analysis). The ROC curve determined sensitivity and specificity of the marker for the detection of stenosis. Results Patients' characteristics were similar between the 2 groups (mean age, 70 years; men, 63%; AVF duration; 39 months), apart from diabetes (control group, 33%; stenosis group, 57%; p=0.04). The rate of OPN was similar between the AVF arm and the contralateral arm (p=0.11), but significantly increased in the stenosis group compared to the control group (655 vs. 452 ng/mL, respectively, p=0.02). There was no statistically significant difference in OPN rate for diabetics and non diabetics (two-way ANOVA, p= 0.50). Sensitivity was 89% for a threshold >293ng/mL and specificity was 80% for a threshold >567ng/mL (AUC: 0.70; 95%CI: 0.57-0.81; p=0.004). Patients with an OPN rate ≥293ng/mL and those with an OPN rate ≥567ng/mL had respectively 8.87 and 15.62 higher odds to have a stenosis than patients with an OPN rate <293ng/mL (binomial regression, p<0.01). Conclusion Plasmatic OPN rate in hemodialysis patients can be measured simply, in a blood sample taken at the time of AVF cannulation, and is strongly correlated to the presence of a symptomatic AVF stenosis. This biomarker could help the physician choose which patient need a comprehensive examination of his/her AVF with duplex ultrasound or fistulogram.